NASA suborbital science and Openscapes proposed support

NASA Earthdata has volumes of remote sensing data collected by platforms in space, air, water, and land. Substantial suborbital data is collected from airborne sampling via airplanes, ground networks, and field sampling on land, boats, and buoys.

Suborbital science teams are teams that create and use suborbital data via NASA programs like EVS-4. The science teams use suborbital data to ground-truth satellite data, as well as to conduct other awesome research like studying Arctic coastlines via the FORTE mission and glaciers and ice sheets via the Snow4Flow mission.

Suborbital teams connect with NASA Distributed Active Archive Center (DAAC) staff at the beginning and the end of the data cycle, and receive additional support from other NASA centers and NASA Headquarters (HQ). The science teams design and run the missions to collect the data using planes, boats, etc, and then archive the data in the DAACs. Then, they also retrieve the data from the DAACs for their analyses, meaning they are relying on the DAACs for the processing as well as storage and distribution. Thus, relationships with science teams and DAACs are very important for effective collaboration and communication, as well as for establishing standards and common workflows.

Staff at many NASA DAACs work with science teams to archive and use suborbital data, including:

• Oak Ridge National Laboratory ORNL,
• National Snow and Ice Data Center NSIDC,
• Atmospheric Science Data Center ASDC,
• Global Hydrometeorology Resource Center GHRC,
• Land Processes LP DAAC,
• Ocean Biology OB.DAAC,
• Physical Oceanography PO.DAAC

Our NASA Openscapes Mentor Community has been collaborating across the DAACs to support Earthdata users and doing so efficiently and with less maintenance burden, while closing previously unseen gaps, together and openly. NASA Openscapes Mentors have created “go-to” resources for staff and users using NASA Earthdata – such as the earthaccess python library and tutorials in the Earthdata Cloud Cookbook. Many other tools and resources have originated from these cross-agency Openscapes relationships with the DAACs, such as VITALS, which has developed specific airborne tools.

In our NASA Openscapes Mentor Community, most of the DAACs listed above that are supporting suborbital work have been consistent contributors over the past five years. Our goal in 2026 is to increase specific support for NASA suborbital science teams to develop reproducible workflows. In March, we are thrilled to welcome Ronny Hernández Mora to the Openscapes core team to support this work! Ronny works at the intersection of Earth data, open and reproducible workflows, and has spent the last past eight years teaching professionals and researchers in industry and academia how to adopt collaborative, open, and maintainable workflows in real-world contexts.

Below is our planned approach from our proposal, presented through POP: Purpose, Outcomes, Process.

Purpose

The purpose is to develop user skills and reproducible workflows for using NASA suborbital data. This work will help communities using NASA Earthdata and open science more broadly, as well as growing the NASA Openscapes community in a new focus area. This work is in collaboration with the CryoCloud (StratusGeo) Community, which is an exciting continuation of the JupyterHub community work we have been collaborating on with them with in the past years.

Outcomes

Suborbital science teams will create reusable open source notebooks (Jupyter, Quarto, GitHub) to analyze their data archived at the NASA DAACs. They can learn no matter where they are starting from, using other existing data or for planning purposes if they do not yet have their own missions data in hand. This may mean learning new skills (coding, version control, documenting and collaborating for reuse), and working with colleagues in new ways as they develop confidence in the open ecosystem to continue their work. At the same time, Openscapes will increase the Mentor capacity of staff at NASA centers by onboarding suborbital-focused user support staff, and also by engaging operational and communications staff, and program managers.

Process

This work will be aligned with NASA HQ priorities to identify and meet suborbital teams, understand goals, progress, tooling, and needs. The following are distinct but overlapping activities: engage • empower • amplify.

1. Engaging suborbital science teams to help build a community. Suborbital teams are already meeting in smaller sub-groups, perhaps based on mission or a specific field; we will strengthen these relationships further as a cross-DAAC community. We will onboard new NASA Openscapes Mentors from NASA’s SoAR (Suborbital Archive Readiness) Team and other parts of NASA to collaborate across the agency to support science teams.
2. Empowering suborbital science teams involves leading a Notebook Clinic that will be remote, cohort-based, and designed for their needs. Lessons will focus on developing reproducible workflows with Jupyter Notebooks, Quarto, and GitHub, using the earthaccess python library, and writing documentation that focuses on reuse and onboarding others to these workflows.
3. Amplifying the work of suborbital science teams involves bringing them into the fold of the NASA Openscapes Flywheel, along with additional NASA center staff. By connecting users (science teams) and NASA Openscapes Mentors and building skills and relationships together, data access issues will be surfaced for NASA development teams (e.g. earthaccess, Cookbook tutorials, cloud optimization).

Cloud Migration and Data Preservation progress happens with collaboration

As NOAA is migrating data to the cloud, this means many people are needing to change their workflows. What Openscapes has learned over the years is that a big part of changing workflows is helping people make connections so they can help each other. Champions Cohorts connect people within and across their home centers as they learn new skills for data workflows that are immediately relevant for their work. And this enables change that wasn’t possible before.

One exciting example of something that wasn’t possible before is the NOAA Fisheries Cloud Computing Setup documentation that NOAA Fisheries Openscapes Mentor Molly Stevens (SEFSC) started in preparation for the Champions Cohort (following six months of collaborative testing with other Mentors). When Alex Norelli learned about it while participating in the Champions Cohort, she found this setup resource valuable for her own work. She began collaborating with Molly to test it, achieving something she could not do before the Cohort started:

Ultimately, Alex co-led a Google Workstations Clinic during the last weeks of the Cohort with Molly and Jon Peake (NMFS Open Science), which benefitted additional participants and helped further develop the documentation. This example embodies how change happens within organizations: people are able to do more together than they can alone. And it starts with the practice of creating something useful for others, sharing it early, and iterating together.

Over two months in fall 2025, 100 NOAA Fisheries staff tackled projects to improve or restructure data workflows through the Openscapes Program. They made substantial progress on complex workflow goals, such as:

• Benchmarking cloud data storage solutions;

• Getting feedback on data modernization policy and needs;

• Developing workflows exclusively in the cloud to align with NOAA Fisheries modernization goals;

• Creating dashboards to communicate data-heavy reports with programmatic code and version control;

• Contributing to coordinated science program onboarding and operating procedures that are harmonized across NOAA Fisheries;

• Exploring how AI can enhance their workflows.

The Openscapes team worked with NOAA Fisheries Openscapes Mentors – staff and affiliates from across centers and offices – to invite colleagues to participate, organize and teach lessons and skill-building workshops, lead small group reflection time, and facilitate coworking sessions. During coworking sessions, participants could brainstorm and make progress on what mattered to them with others working on similar tasks.

Fall 2025 Cohorts - what was new

Adapting our standard Champions Cohorts, we made more space for NOAA Fisheries Openscapes Mentors and NOAA leadership to share new tools and policies for cloud migration.

• Michael Liddel (OST - Assistant Chief Data Officer) and Heather Nicholas (Office of the CIO) hosted Seaside Chats on Cloud migration and Data Optimization while participating as Champions and learning about science staff needs.

• Eli Holmes (NMFS Open Science) developed and taught a new lesson on Cloud Strategies for Future Us (recording).

• Molly Stevens (SEFSC), Alex Norelli (SEFSC new Champion), and Jon Peake (NMFS Open Science) documented the NOAA Fisheries Cloud Computing Setup and developed and hosted a Google Workstations Clinic.

• Kathryn Doering (OST) presented on Infrastructure and support at NOAA Fisheries.

• Erin Steiner (NWFSC/OST) launched and presented on the new GitHub server for confidential information now available across NOAA Fisheries.

We also partnered with Intertidal Agency’s Kate Wing and Rachael Blake to teach new lessons to address the theme of data preservation: Metadata - Documenting your Data and a Zenodo Clinic. These were built from an ESIP summer conference session “Archive your first or second dataset”. This is a powerful partnership; Intertidal is leading a Data Stewardship Cohort shortly (registration open!) and will reuse the Openscapes Champions structure and some of these lessons. The Zenodo Clinic replaced our regular GitHub Clinic. However, the GitHub Clinic is something people asked for, as it has been recognized by many as a friendly on-ramp to GitHub concepts and collaboration. We taught aspects of the GitHub Clinic in three coworking sessions with small groups that included both beginners and more experienced users.

What NOAA Fisheries staff accomplished: learning and shifting workflows

In the final cohort call, people are invited to share their work-in-progress using our Pathways tool for identifying how they work now along with their goals and progress.

We saw several themes popping up:

• asking for help;

• “slowing down to speed up”;

• seeing what’s possible;

• testing things together - like cloud infrastructure - checking in with each other during seaside chats, trying to make things work, comparing workflows;

• sharing templates for workflows, like project management approaches for cyclical stock assessment reports.

We also saw that many more people participated as individuals rather than in teams this time including some who are collaborating with IT colleagues at several centers. This is why the experience of “learning all of the faces; these are people doing the same work I’m doing” and knowing who to ask for help is so valuable, while often underappreciated as an impact of cohort-based learning.

Here are some examples from those presentations during our final call together.

Mentors impact – building from over 5 years of collaboration

After five years of working with NOAA Fisheries, we’re seeing Openscapes practices really become a movement. People are learning from each other (“now we see that our team leads were ‘Openscapesing’ us all along with their open process!”). Champions are becoming Mentors and turning around to share with new Champions (like the project management examples above) and inside their divisions as they see needs, and mentoring through seaside chats, building, and sharing to address those needs. We’re seeing the Openscapes approach reach beyond the early adopter innovators to the early majority (more on this under Reflections, below).

The impacts we see now build on work that develops over years. The Google Workstations Clinic that Molly Stevens, Alex Norelli, and Jon Peake delivered stemmed from a MentorActivities Issue on the topic from early 2025, and Molly ran with it. That first issue has turned into a huge and ongoing impact across NOAA Fisheries.

It shows up in other ways too. Loren Stearman (NWFSC) is focusing on psychological safety with the teams he works most closely with. He is working on how to help bring Openscapes principles beyond data workflow contexts. The human side and safe environments for communication is so critical to the longevity of the projects he works on and he is helping other Mentors across NOAA Fisheries prioritize and have language for these conversations.

Perseverance and envisioning what’s possible is another characteristic of NOAA Mentors. Erin Steiner (NWFSC) saw the need for a GitHub server for confidential data - losing a system her local team had relied on meant she could rebuild it for everyone! Erin developed a plan and proposal, and did not give up, spending time in an 8-month journey convincing people that there was both a need and a solution. She shared this as a Pathway for the 2025 fall Champions cohort. Erin says the hardest part was not quitting, even when she was told she was using GitHub wrong (she was not!). It is hard to know what was different between the 58th version of a pitch email and 59th, but it was that 59th email that gained traction to make it happen. Erin gave a shoutout to some IT infrastructure folks. And ultimately, she was successful (the beta version launched Feb 2nd!) because she was not doing this alone: she was able to reuse (“fork”) the approach Kathryn Doering (OST) took for GitHub Governance (seeing what’s possible), and had the encouragement of Kathryn and others along the way.

Reflections: Crossing the Chasm with Innovators and Early Majority

This year’s NOAA Fisheries Openscapes Champions program was different in two big ways that were happening simultaneously. We are reaching new audiences, having worked with NOAA Fisheries for so many years, and led 16 Champions Cohorts with them. Thinking about the diffusion of innovation theory; for the past years we have been investing in the “early adopter” community by supporting the NOAA Fisheries Mentors and Champions cohorts. In 2025, we focused on sharing the story of how we had “crossed the chasm” and would now be working more with the “early majority”. The Fall 2025 Champions Cohort was the next part of the story: we had early majority folks as well as innovators who are developing new infrastructure and workflows for cloud computing.

What does this mean for how we engage and teach NOAA Fisheries Champions going forward? Will the spectrum of NOAA staff’s goals have broadened in a way that the support we give people will look different in the future? We’ve always designed Champions Cohorts to not assume prior knowledge of coding or GitHub or any specific dataset or type, and we’ve designed so that people with different expertise and deliverables, like supervisors and analysts and IT staff, can all learn together and take what they need back to their work. However, we need to think more about what Openscapes Champions looks like when the spirit of “data workflows” is not a unifier for all participants. While we will need to listen and innovate to meet these new needs, we will also remember to pause and reuse what has worked in the past: the value of welcome, art, show-not-tell, empathy, and community. And a constant reminder that we’re not alone, it’s not too late.

Goals

The goal we pitched to ESIP’s FUNding Friday was to build a prototype and roadmap for “Growing the Family” by integrating earthaccess with non-NASA Earth science data (NOAA, USGS, Copernicus, and more!). We designed the event to focus on building the community of earthaccess contributors and adopters; we were equally interested in growing ideas as much as we were interested in developing technology to support earthaccess growth. For those new to earthaccess, it’s a Python library that simplifies access to NASA Earth science data, turning complex authentication and data discovery into just a few lines of code. It’s becoming essential infrastructure for researchers working with Earth observation data, and with this hackday, we wanted to expand its community and expand its reach beyond NASA to other Earth science data providers.

Zoomie (Zoom selfie) of virtual and in-person hackday participants (with earthaccess hex stickers!).

The hackday: what we did

The hackday was hybrid: online and in-person in New Orleans (independent of AGU2025 but conveniently co-located for attendees). We shared four tracks in advance as topics folks might want to focus on:

• STAC (SpatioTemporal Asset Catalogs) Integrations – “Growing the Family”

• User Stories – Understanding Pain Points and Priorities

• User Onboarding and LLM Support

• Data Virtualization

Leads opened GitHub Issues or Discussions for tracks so that folks could consider and discuss these in advance. This is part of lowering barriers to participation, but also capturing progress, notes, and discussions.

We had a Zoom breakout room for each track (everyone was logged into Zoom to be more inclusive of remote participants). We had one Agenda document (screenshot) with Tabs for each track, and time at the beginning, middle, and end for all tracks to come back together in the Main room and share out.

Planning the hackday

Planning the hackday was a study in how generous people fuel collaboration across groups, reusing (“forking”) existing structures and using existing communication channels, and being comfortable with “good enough” given the time and resources we had. For example, some folks were only able to lead during a portion of the session time, and many were not able to fully polish slidedecks ahead of time. Our event planning phase was during the government shutdown, so we were missing people, and others were over-taxed with responsibilities. Even so, folks prioritized time where they could, and we focused on communicating openly and early to invite participants to sign up and also to coordinate our facilitator team and design the plan. CNG created a registration page for participant sign-ups that we shared with our communities. We coordinated largely through a single spreadsheet where we listed “hackday facilitators” (earthaccess maintainers and NASA Openscapes Mentor community) and mapped out availability and roles.

Hackday funds from ESIP went to the venue rentals for the hackday and social hour, smoothly coordinated by CNG. Hackdays cost more than $5K, however, and this did not compensate for the time for the 22 facilitators involved. Their time was generously covered “in kind” by their employers (listed in the preface).

We have survey responses that will help us support earthaccess adopters and contributors – and help us lead successful events in the future.

Building Community, not just Code

There is so much value in the people and organizations working together to lead this event, it’s impossible to overstate. We saw our approach have impact: building a welcoming contributor community (users are contributors!) around a transformative python library supported by technical and human infrastructure. Diverse participants, some who’ve already been involved and some new to earthaccess who were keen to collaborate came from many different organizations, including NASA DAACs, DevelopmentSeed, UCBerkeley, Earthscope hub, and NOAA. They were developers, contributors, newcomers, users, learners. People joined from across the US, and from Poland, India, and Canada.

Several people indicated they were new to earthaccess and for some this was their first-ever hackday! We aimed to “grow the family” and ease people’s introduction by running a 2-day event in November - A welcome to NASA Earthdata and earthaccess. Day 1 consisted of lessons and demos and Day 2 had people getting hands-on experience by running those demos themselves in breakout rooms they chose. This clearly lowered some barriers to hackday participation and contribution to earthaccess in general. One participant joined the November event, where she got more familiar with contributing, joined our biweekly earthaccess hackdays, and then contributed in this hackday. She’s now a full contributor with pull requests merged into earthaccess!

Early Outcomes and next steps ESIP Summer Meeting 2026

We continue to have Tuesday biweekly hackdays to move progress forward - please join us! - and we’ll continue on these Tracks as well. To learn more, please peruse the discussions, issues, and other content on the earthaccess GitHub repository.

Here are some early outcomes (many still being worked on!), that we will describe in more detail at the ESIP Summer Meeting:

STAC Track:

• GitHub Issues; Discussions.

• We created a Pull Request (https://github.com/nsidc/earthaccess/pull/1171)  that’s a test of querying CMR (Common Metadata Repository) and parsing them as STAC (SpatioTemporal Asset Catalogs) items.

• Most important: Deciding not to reimplement pystac client.

• End goal to have same 3-lines of code to access anything with a STAC client. Maybe more modular.

User stories Track:

• Focusing around engagement of earthaccess. How do users find earthaccess? How do they initially engage with it? What are some pain points?

• Ideas to ease engagement: Creating a zine and cheatsheet, widely used in the R community

• Next steps: Ideas around what to include in cheatsheet(s), how to expand reach, and how to make earthaccess more accessible for more people

User Onboarding Track:

• Leading newcomers through basic tutorials, including OPeNDAP tutorial that Chris Battisto ran

• Welcoming newcomers, experimenting accessing Daymet data that a participant works with

• Good questions around services and how it relates to earthaccess (ex harmony, harmony-py and OPeNDAP)

Data Virtualization Track:

• First half focused on shared learning about virtual datacube concepts and implementations

• Lots of next steps:

  • DevSeed/Aimee Barciauskas spinning up monthly meeting on Virtual Datasets beginning January 2026

  • James Gallagher is organizing a session at ESIP to come up with a standard for chunk manifests… virtualizarr, zarr, dmrpp, and more, with a goal to have a draft for the summer meeting

  • Max Jones posted some cool metrics on how fast data access can be with virtual data cubes with different stacks (Luis)

  • Danny Kaufman was inspired to finally work on Icechunk-ifying a collection from ASDC!

Learn more about our work with our partners since our last newsletter

Sophie Breitbart: Cheatsheet for the asar R package

At a late August NOAA Fisheries Openscapes coworking, I shared that I’d finished the first version of the {asar} cheatsheet. After some further feedback, the cheatsheet was updated once more. We plan to continually update the cheatsheet as our package evolves. Here’s some insight into how I made this cheatsheet (my first!):

Samantha Schiano: Cheatsheet for the stockplotr R package

I am a member of the team that develops {asar} (the package mentioned by Sophie Breitbart earlier in this blog post!). We also develop a sister package called stockplotr which focuses on creating tables and plots for fisheries stock assessments. There are an increasing number of awesome functions we have developed to allow users to create their plots and tables in one click or customize themselves. While Sophie took on the task of developing the cheatsheet for {asar}, I took on creating the cheatsheet for {stockplotr}. I aimed to make a similar looking cheatsheet to {asar} in order to create a cohesive feel across our packages. It was also my first time making a cheatsheet and I learned a lot from Sophie and through my own process.

Kelli Johnson: Cheatsheet for the FIMS package

Like others, my inspiration for creating a cheatsheet came from my consistent use of Posit cheatsheets. In fact, I have my book of cheatsheets from the 2024 Posit conference within arms reach of my desk at all times. I find that these cheatsheets offer information that just cannot be found through the traditional R documentation, i.e.,?function_name, and I often go to them for their visuals rather than their text.

Official AGU Citation - Greg Leptoukh Lecture Award

Note: this citation and response will be posted on Julie’s AGU profile in the coming months.

Dr. Julia Stewart Lowndes is transforming environmental and Earth science through her visionary leadership in open, collaborative, and inclusive data science. She has redefined how scientific teams work together—empowering researchers to do better science in less time, together—and inspiring a movement toward more effective, inclusive, and reproducible Research.

Trained as a marine scientist with a Ph.D. from Stanford University, and later as a scientist at the National Center for Ecological Analysis and Synthesis (NCEAS), Dr. Lowndes gained key insights into the challenges faced by solo scientists when working with big data. Her 2017 paper in Nature Ecology and Evolution, “Our path to better science in less time using open data science tools,” (Lowndes et al. 2017) articulated a compelling vision for modernizing scientific workflows. By addressing both the technical and human challenges of data science, her work has catalyzed cultural change across government agencies, academic institutions, and research networks.

In 2018, Dr. Lowndes founded Openscapes, a pioneering program that mentors individuals and research teams in open, reproducible, and collaborative science. Through initiatives such as NASA Openscapes, she has reshaped how NASA data centers and researchers collaborate—accelerating cloud migration, strengthening interdisciplinary connections, and fostering psychological safety within teams. Openscapes’ work with NOAA, NASA, and EPA has advanced the culture of open science across the federal research landscape and was featured in the White House’s 2024 Year of Open Science fact sheet for its exemplary mentorship with NOAA Fisheries.

Dr. Lowndes is also a dedicated advocate for equity and inclusion in data science. She co- founded R-Ladies Santa Barbara and advises the Pathways to Open Science program supporting Black environmental and marine researchers. Her efforts have built welcoming communities that empower those historically excluded from technical fields to lead in open Science.

Dr. Lowndes exemplifies the spirit of the Greg Leptoukh Lecture, which honors innovations in informatics that enable scientific discovery. Like Greg Leptoukh, she recognizes that progress in science depends not only on technical excellence but also on collaboration, transparency, and community. Her leadership has created a more inclusive, efficient, and hopeful model for how science is done—making her a profoundly deserving recipient of this honor.

Aaron Friesz, Earth Science Information Partners (ESIP)
NASA Openscapes Mentors and Allison Horst

Julia’s Response - Greg Leptoukh Lecture Award

It is an absolute honor to receive the Greg Leptoukh Lecture award recognizing innovations in computational and data informatics that enable scientific discovery. I feel an intense upwelling of love and gratitude for my colleagues that is difficult to describe. It is a privilege to feel such joy about one’s colleagues. And yet, it is quite a common trait of our greater open science community that reflects the deep value of collaboration through trust, empathy, and kindness.

I created Openscapes to be a feeling of welcome and empowerment around data-intensive science, having felt this myself as a marine ecologist welcomed and empowered by the R open source programming community. Since 2018 we’ve been directly mentoring nearly one thousand researchers and people supporting research, while co-creating mechanisms together so we can all turn around and support countless others within our work-lives. We continue to learn from, with, and for each other, every effort turning our Flywheel (Robinson and Lowndes 2022). We reuse and “fork” what works in new places, using a concept from software engineering for reuse and credit. Openscapes now means more to people than I can possibly define. Its impacts are impossible to quantify, as so many “n of 1” individuals make a whole wave of change in a different place, and in a way that is specific and empathetic to that place. This is movement building. The connections and innovations across interleaving communities and what it means for science and society is beautiful.

My nomination for the Leptoukh Lecture is one example. My dear colleagues Amy Steiker and Luis Lopez of NSIDC (National Snow and Ice Data Center) – representing the NASA Openscapes Mentors co-creating and teaching common tutorials to support researchers as they migrate analytical workflows to the Cloud – collaborated even as Aaron Friesz moved on to a new job at ESIP, where he supports the data community in new ways. And they included Allison Horst, a data scientist, teacher, and longtime collaborator who developed all Openscapes art, but who they had not yet met. They forked their systems for co-development and drew from our open community networks for a new purpose, just as we do when we build other social and technical infrastructure for science.

We’ve got more work to do. As Johnson & Wilkinson write in “All We Can Save” continue to drive us: “We” speaks to the collective, to collaboration, to community, to the relational work at hand. Addressing the climate crisis…will take everyone. We cannot, we must not, go it alone.”

Dr. Julia Stewart Lowndes
Openscapes
Founding director

Navigating NASA Earthdata together with earthaccess

NASA Earthdata has a 5-year track record of successful and collaborative Cloud trainings. NASA Openscapes Mentors are supporting Earthdata users and doing so efficiently and with less maintenance burden, while solving previously unseen gaps, together and openly. This was evident in the way we were able to lead Champions by leveraging our existing systems and reusing materials during a time when many NASA staff were furloughed. For our Nov 13-14 events, earthaccess contributors from 4 different NASA Data Centers (NSIDC, ORNL, ASDC, PO.DAAC), as well as Development Seed, University of New Hampshire, and Openscapes collaborated and co-taught using resources co-developed by the NASA Openscapes and earthaccess communities. See the slides for more details on each topic: 

• What is Openscapes & NASA Openscapes?, Julie Lowndes (Openscapes) + Amy Steiker (NSIDC)

• NASA Earthdata & cloud computing, Michele Thornton (ORNL)

• Accelerating access through open, collaborative dev, Amy Steiker (NSIDC)

• Using earthaccess - quickstart & in action, Mikala Beig (NSIDC)

• Streamlining analyses for real research projects. ex ICESat-2 tracks with Mosaic Cruise (temporal) (notebook) (20 min) - Andy Barrett (NSIDC)

• Vision for earthaccess: Broadening the scope & virtual datasets Danny Kaufman (ASDC)

• Resources and Contributing - Cookbook and Open Communities, Celia Ou (PO.DAAC)

The goals of the events were to provide a friendly welcome to NASA Earthdata and earthaccess for new and seasoned adopters*. We wanted participants to feel ready to get hands-on experience in Day 2; feel comfortable using earthaccess; provide the tools to engage in the community and to contribute; and capture needs and feedback. Here are a few things that stood out from these events.

*Credit to Max Jones (Development Seed) for using the word “adopters” rather than “users” in describing the recent Zarr Summit.

Continuing in Winter 2026

We will continue the NASA Openscapes Champions Program in 2026. In 2025-2026, we are trying something different. We are modifying the traditional Champions structure to be more light-weight so that participants can engage at whatever level they can (rather than the full 10-week cohort-based model of usual Champions programs). Following the Fall 2025 events, in Winter 2026 we will lead a 4-Part Series of Community Calls and Coworking sessions, likely February - April. With the disruptions from the US government shutdown, we are delayed in our planning, and will announce the dates and plan as we develop it at https://nasa-openscapes.github.io/champions.

A peek at the sessions Openscapes co-led:

• Consistent, reproducible workflows for geospatial stewardship, with Rich Signell, Julia Lowndes, Brianna Pagán, Kate Wing, Carl Boettiger, Jed Sundwall. This session aimed to create consistent, reproducible workflows for others to learn and use.

• Crossing the chasm: amplifying success stories about co-creating across institutions, with Julie Lowndes, Eli Holmes, Jennifer Schopf, Yuvi. (slides). This session shared four stories of the Diffusion of Innovation theory in practice: at NOAA Fisheries, at NASA Earthdata 2021-2022, at the Texas Advanced Computing Center by saving 50 years of astronomy data with Arecibo, and Finding Chasms to Cross: JupyterHubs at Scale from Wikimedia to UC Berkeley to your Research Institution.

  • Note: since NASA Mentors were denied attendance, we prepared a whole Community Call for them to share the following week: earthaccess — helping users leverage the awesomeness of NASA Earthdata

• Archive your first or second data set, with Julie Lowndes, Joseph Gum, Eli Holmes, Kate Wing, Rachael Blake. (slides) This was a hands-on companion session to “Case Studies of Transitioning Datasets from Specialist to Generalist Repositories”. The vision here was be to help people figure out where and how to archive their first or second data set, using Zenodo as an explicit example.

• Unconf to prepare for the NASEM’s Future Directions for Earth Observations and Data Stewardship conference in December (details & registration), with Steve Diggs, Kate Wing, Rachael Blake.

• FUNding Friday pitch for an earthaccess hackday - this was successful!

earthaccess hackday on Dec 15!

We are hosting an earthaccess hackday on Dec 15! The goal is to build a prototype and roadmap for “Growing the Family” by integrating earthaccess with non-NASA Earth science data (NOAA, USGS, Copernicus, and more!). This is supported through the ESIP’s FUNding Friday initiative, and is co-hosted by Cloud-Native Geospatial Forum (CNG) and Openscapes. Please join us! Details & registration.

ESIP’s FUNding Friday is a way to pitch an idea and win $5K for an idea. Participants from the conference make posters the evening before at a social hour. You can read about this in ESIP’s FUNding Friday blog post. In Seattle 2025, we won! We pitched an idea through song; lyrics below, along with a 1.5 min (video), and then summarized:

earthaccess is a Python library that enables authentication, search, and access for NASA Earth science data with just a few lines of code. We see great potential for growing the family to include data from NOAA, USGS, and Copernicus in Europe. We would use FUNding Friday funds for a hackday, virtual and in person, at AGU in December. We expect a workable prototype and roadmap in an afternoon. Thank you.

What follows in the timeline is a way to document what we did, and to credit non-traditional contributors and contributions. Putting our pitch together was a joyful collaboration with contributors from NASA, NOAA, Intertidal Agency, Development Seed, OPeNDAP, 2i2c, Openscapes, and passers-by. This would absolutely not have happened without hours and years of prior connections, collaborations, trust- and community-building, and technical and social interactions among the four presenters and our collective web of smart, caring humans.

Posts about our ongoing Water Boards - Openscapes collaboration

California Water Boards Mentor Perspectives, 2021

3 takeaways for planning for the year of open science, 2022

Adapting the Champions Program for the California Water Boards, 2022

3 approaches for the year of open science, 2023

California Water Boards’ 2nd Annual Openscapes Champions Program - Reflections & Future, 2023

Shifting institutional culture to develop climate solutions with Open Science, 2024

Participants learned to

• Recognize how image classification of airborne imaging spectroscopy and labeled field data can be used to map invasive species.

• Use a Jupyter Notebook to access NASA JPL AVIRIS-NG data located on the NASA Earthdata cloud for a spatial and temporal region of interest.

• Apply a cloud based workflow in a provided Jupyter Notebook to classify species using a trained machine learning model for the Greater Cape Floristic Region, South Africa.

• Identify key considerations for interpreting machine learning end products for invasive species mapping for land management decisions.

Target Audience

• Resource managers (local, state, regional, international), agriculture sector, remote sensing technicians, ecologists, and academics.

Hands-on exercises were executed in a JupyterHub on the Openscapes 2i2c NASA workshop cloud instance. Using the cloud-hosted JupyterHub for the workshop allowed participants real time and active participation during the Notebook presentation. This is significant in that it allows global participants successful access to NASA Earthdata resources and demonstrates the processes of discovery, access, and analysis of NASA Cloud-hosted Earthdata data. Participants were allowed access the the workshop hub for an extended period of two weeks to continue their exploration of the Notebook and resources.

We presented the following notebooks from the ORNL DAAC’s repository of NASA Airborne Data Science Tutorials:

• Mapping Invasive Species Using Supervised Machine Learning and AVIRIS-NG

Lessons learned

• 2i2c was amazing in allowing for an extended period of Workshop Hub access and juggling a few workshop needs over that time period

• The training video will continue to be public. We removed the username/password information from the recording, but we did not provide content for running Notebooks outside the hub for future users.

Crossing the Chasm: The Journey Toward Reproducibility

A recurring theme throughout the day was the ongoing transition from legacy, siloed workflows to open, reproducible, and cloud-compatible practices. These innovations, while forward-moving, are at times disruptive. The plenary provided a time for reflecting on and sharing lessons learned from across NOAA Fisheries about these transitions.

During the first half of the plenary session, Eli Holmes, Lead of NOAA Fisheries Open Science, provided a set of concepts on organizational change: diffusion of innovation and the adaptation of that work by Geoffrey Moore, crossing the chasm. Eli viewed these ideas as a set of tools, alongside GitHub or other software tools, that help her understand and explain her own work. She demonstrated this approach by reflecting on the evolution from her own decision to learn to use GitHub to its adoption across NMFS today.

As Eli and a few initial colleagues incorporated GitHub into their own workflows for sharing models, they began to envision broad use across NMFS. Eli likened this early GitHub work to the first person who thought a wolf could be trained to herd sheep—a wild, disruptive idea at first, met with skepticism from those focused on raising sheep, not training puppies.

Most people in an organization would not consider themselves eager to train a puppy or dramatically change their workflows. This is the chasm between early adopters and everyone else. Eli emphasized that for an idea to diffuse into the larger organization, early adopters must be aware that there is a chasm. Being aware of the chasm and that it is a normal part of these kinds of change can reduce frustration in the early adopters, who often wonder why others aren’t as eager to make a change as they are.

Being aware is not enough. Crossing the chasm requires the initial ideas to be shaped into something that solves a real problem for the community and is practical and easy to adopt. Continuing with the sheepdog metaphor, early adopters put in the patience and obsession to turn that “puppy” into a skilled sheepdog. Looking back, Eli highlighted how those early meetings and experiments with a small group of early adopters, and slow progress over time, were the essential work required to cross the chasm and support and move from eager adopters to those more hesitant to change.

Ultimately, NMFS successfully crossed the GitHub adoption chasm. The NMFS’s early GitHub experiments evolved into GitHub Enterprise licences that are available equitably across centers, processes are documented, and a GitHub Governance team is established at NOAA Fisheries. This retrospective anecdote provided encouragement to those among us who are innovators and early adopters. We hope that some might take these concepts and ‘fork them’ to understand, troubleshoot, and communicate other technology transitions.

In the second half of the plenary, mentors used the Moments, Memories, Meaning approach to eliciting stories, to highlight progress through NMFS Openscapes. One mentor shared experiences of the lag between having the tools available and gaining traction with their team to use them. Another mentor described how automating stock assessment reports, one section at a time, over multiple years, eventually freed up time to focus not just on delivering outputs, but improving the workflow behind them. A third mentor knew they had reached a tipping point when it was easier to do things the new way, in Quarto, than the old PowerPoint way.

Without documentation, automation, and shared protocols, scientific work remains stuck in desktop-bound silos. But with them, it becomes portable, scalable, and future-ready. The lesson from these stories is clear: transformation takes time. But each script shared, repo forked, and quarto book published is a step toward workflows that don’t just work locally—they work across teams, in the cloud, and for the long haul.

Project Progress During Co-Creation Blocks

During the morning and afternoon co-creation blocks, participants worked on active projects with the support of Openscapes mentors and peers. These sessions were a twist on the agile format stand-ups. Participants shared their problems quickly, the entire audience asked clarifying questions, and then a small group had a short breakout on each problem to identify more resources, advice, or questions. With this quickly crowdsourced information, the teams interested in working on each project then split off for dedicated time to make tangible progress.

Below is a list of projects that were worked on during the workshop:

1. Centralizing NMFS Branding Assets
Participants worked on centralizing NMFS branding assets for use across scientific outputs. This included referencing theNOAA-NMFS-Brand-Resources GitHub repository, which houses reproducible elements like colors, fonts, and layout guidance. The goal is to support consistency across reports, websites, and Quarto projects.

2. Decision Tree for Moving On-Prem Data to the Cloud
This group focused on outlining when and how to move datasets from on-premise storage to the cloud. The resulting decision tree is being developed as part of the NMFS Cloud Data Resource Book. This project was identified as foundational to cloud readiness.

3. Research Compendium for PIFSC Marine Turtle Camps and Beyond
This team focused on adapting a research compendium template to the needs of marine turtle fieldwork and other data-collection-intensive projects. Their work builds on existing compendium structures to enable reuse and reproducibility across related efforts.

4. Applying Quarto for Automating and Templating Reports
This project explored how Quarto can be used to generate repeatable report templates. The group investigated how to streamline authoring scientific reports across Centers by using parameterized templates and integrating them into version-controlled workflows. An example workflow was tested during the co-creation time. This work builds on the already advanced-level Quarto applications developed by the mentors and their colleagues.

5. Open Science Tools Decision Tree
This group worked on developing guidance around which tools to use (e.g., GitHub, Quarto, RStudio) for different stages of scientific work. Their goal was to make open science workflows easier to adopt, particularly for those new to coding or version control.

6. Central Locations for Open Science Tools at PIFSC
This team focused on designing a more discoverable and centralized place to access open science resources within PIFSC. Their intent was to “fork” the model used by other groups to curate tools and templates for broader visibility and reuse across the center.

7. “Code Cleanup” of R packages in the National Stock Assessment Program
The group who maintains or contributes to the centralized asar and stockplotr R packages for automated stock assessment reports used the time for a collaborative code cleanup.

What Comes Next: Enabling NOAA’s Cloud Transition

Moving into the next year of work, cloud migration is a central thread to carry forward. Participants emphasized that being “cloud ready” requires more than storage space—it depends on well-documented, reproducible workflows that enable others to understand, reuse, and trust the work. Through the co-creation blocks, they identified actionable gaps in training and decision-making, underscoring the need to invest in collaboration-centered approaches to technical change.

Key needs emerged as NOAA Fisheries looks ahead:

• Build pathways, not just tools, especially for complex workflows involving cloud storage and access

• Tailor documentation and training to NOAA scientists, with their real-world collaboration and compliance needs in mind

• Broaden the mentor community to include data stewards, analysts, managers, and IT staff

• Amplify successful projects and template solutions across centers

• Acknowledge and support the “invisible work” of documentation, onboarding, and peer coaching

From the many moments shared over the course of the day, a real shift occurs when reproducible workflows are not just better—they’re the obvious choice. The workshop highlighted how NMFS Openscapes is enabling NOAA Fisheries to lead by example—putting core scientific values into practice through open workflows, shared learning, and intentional collaboration. 

Crossing the Chasm at NASA Earthdata with earthaccess

Amy Steiker started off by sharing about the initial chasm between users and Earthdata Cloud, using the “Crossing the Chasm” framing in relation to the earthaccess origins. Eli Holmes (NOAA Fisheries) led the Crossing the Chasm session at ESIP last week, presenting how ideas spread through an organization and the challenging gap between early adopters and broader adoption. Amy shared how earthaccess was built as the bridge to support users adopting cloud workflows, enabling seamless search, authentication, and access (Figure below).

Amy shared the impact of earthaccess using statistics on usage and contributions from the open development community. And while there are new challenges in the face of budget uncertainties and a lack of sustained funding, the intentional culture and community building remains critical to the library’s success. We can now flip the question of how to support earthaccess, and instead ask: How can earthaccess support NASA Earthdata? Amy concluded by sharing a vision we’re building of “earthaccess with batteries included” (credit to Andy Barrett, NSIDC) and “Widening the bridge”, inviting contributors shape the future!

Amy emphasized how partnerships are critical. She shared how earthaccess won the ESIP FUNding Friday $5k, an annual mini-competition awarding small projects that are inspired by ESIP collaboration or participation. We plan to use funds to host an earthaccess hackathon at the AGU December meeting - virtually and in person! (Here is the FUNding Friday video pitch 🥳🎶, illustrative of the social infrastructure underpinning how we work. We will continue telling this story, and also trying to document the timeline getting there).

New features and VirtualiZarr

Joe Kennedy’s presentation on virtual datasets focused on how cloud computing paradigms enable small teams to generate near real-time mission-scale derived data products. He showed how virtual datasets are used by the ITS_LIVE project to provide low-latency access to global glacier velocity data from multiple optical, radar, and laser satellite sensors. This is achieved by extracting time series of measurements from a particular location and computing velocities by measuring the distance moved over time. Zarr datasets support this targeted extraction natively. Virtual datasets make this possible for other data storage formats by utilizing metadata (via the kerchunk python library) that allows extraction of data for a particular location and time, rather than having to process the entire granules.

Joe put forward a vision where earthaccess can generalize this access pattern for all datasets. This work is underway through integrating VirtualiZarr and Icechunk into earthaccess. He emphasized critical challenges in implementing virtual datasets effectively. The core issue is that cloud optimization isn’t just about file formats - it’s fundamentally about aligning data organization with data access patterns. Data providers face difficult trade-offs: chunking data temporally optimizes time series analysis but degrades spatial access performance, and vice versa. Additional complexities include handling authentication, and navigating institutional policies around data egress and regional deployments. Joe stressed that success requires alignment between data providers, services, and community adoption.

Deeper specifics with VirtualiZarr

Chris Battisto talked about updating and managing VirtualiZarr stores using Icechunk. He highlighted two main approaches for creating Earth data cubes using xarray with NASA Earth data: direct Zarr store access with spatial/temporal subsetting (which requires being in AWS US-West-2), and individual granule download and concatenation (common but memory and storage-bound).

He then explained a third way: Create virtualized datasets with VirtualiZarr, and store (and update) with Icechunk. While virtualized datasets offer significant advantages in terms of scalability and efficiency, Chris noted there are still challenges in getting them to work with all NASA Earth data collections. He encouraged the community to experiment with these tools and share their experiences through the earthaccess GitHub repository, and provided documentation links for getting started with both VirtualiZarr and Icechunk implementations.

Next, Danny Kaufman presented on using virtual datasets with TEMPO (Tropospheric Emissions: Monitoring of Pollution) data. TEMPO monitors air pollution over North America and generates substantial data volumes: approximately 10 Level-2 (L2 - non-gridded) granules per hour and about 5,000 Level-3 (L3 - gridded) granules (2.5 TB)  annually. Danny found that L3 data products worked well with earthaccess’s virtual dataset implementation, which substantially reduced data access time from about one day when downloading to just ~10 minutes. This dramatic improvement was possible because L3 TEMPO data has both consistent dimension sizes and pre-generated DMR++ reference files, which contain chunk manifests with byte ranges for efficient S3 access. In contrast, L2 TEMPO data failed to work as virtual datasets because of inconsistent dimension sizes across granules, which isn’t compatible with current Zarr specifications. 

Danny showed that roughly 15% (given several caveats) of NASA Earthdata collections have the necessary DMR++ files that are both downloadable and parseable by VirtualiZarr. To further streamline access, the team is working on creating combined reference files that don’t rely on DMR++ for the entire Level 3 TEMPO collection. As with all of these initiatives, Danny emphasized that documentation and community learning remain crucial for broader adoption.

Bringing It Home

Luis Lopez brought us home with a presentation focused on earthaccess’s role as a bridge between the current state of NASA Earth Data and a future cloud-native, analysis-ready archive. He emphasized that earthaccess serves as a crucial abstraction layer that helps scientists navigate the complexities of NASA’s data systems. Currently, the library is working to implement smart data access patterns, particularly through virtual data cubes, to enable efficient scientific analysis whether users need time series data or multi-file operations. The immediate technical priorities include handling distributed authentication, optimizing metadata reads, and improving I/O operations.

Looking forward, Luis highlighted a fundamental shift in thinking about NASA’s data challenges, quoting Patrick Quinn: “Stop Building Tools. Start Fixing Data.”. He stressed that data should be harmonized and optimized at the source, which would reduce the need for numerous APIs and workarounds. The overarching mission for earthaccess is to continue bridging the gap between the current fragmented state of NASA data archives and an ideal future where data can be accessed with the same simple three lines of code regardless of the underlying complexity. This vision aligns with the broader goal of preserving simplicity for users while handling the technical challenges behind the scenes.

Resources

• To join as a contributor: https://earthaccess.readthedocs.io/en/latest/contributing 

• earthaccess documentation 

• earthaccess: Earth Science Data Simplified

• Earthmover documentation 

• From Chris Battisto’s slides

  • VirtualiZarr Docs

    • Usage Questions

    • Data Structures

    • End-to-end Examples

  • earthaccess Issues

  • Icechunk Docs

    • Managing Repos

    • Arraylake

Speakers

• Amy Steiker, National Snow and Ice Data Center (NSIDC) Distributed Active Archive Center (DAAC). Amy specializes in the usability and accessibility of NASA Earthdata. She has been a NASA Openscapes Mentor since 2021, working to foster the growth and sustainment of the earthaccess community.

• Luis Lopez, NSIDC DAAC, is a Research Software Engineer at the National Snow and Ice Data Center (NSIDC) in Boulder, Colorado. He has helped develop tools and services to facilitate data access and discoverability across different NASA Earth missions. He is a passionate advocate of open science and has contributed to open source projects such as Apache Nutch, PyLDAVis and others. He’s always happy to help scientists find ways to make their workflows more efficient. Luis has presented his work at SciPy LATAM, PyData Global and the IEEE annual Big Data symposium.

• Danny Kaufman, Atmospheric Science Data Center (ASDC) DAAC. Danny’s background includes ocean and climate data analyses, development of decision support tools for pollution management and data pipelines for natural language processing. He currently supports ASDC with the TEMPO air quality mission, with NASA’s Open Science initiative, and as a NASA Openscapes Mentor since 2023 — co-developing and delivering tools and learning materials to facilitate user-friendly access to NASA Earthdata.

• Joe Kennedy, Alaska Satellite Facility (ASF) DAAC is a Staff Scientist at the Alaska Satellite Facility, a Computational Glaciologist, a Climate Scientist, and an Adventurer living/playing in Alaska ⛄. He develops (primarily) Python-based, open-source scientific software and focuses on scientific workflows that scale from laptops to HPCs or the Cloud.

• Chris Battisto, Goddard Earth Sciences Data and Information Services Center (GES DISC) DAAC. Chris Battisto is a senior scientific developer and a member of the GES DISC User Needs team, working to curate and contribute tutorials that connect scientists of all experience levels with NASA Earthdata across several programmatic tools. He regularly collaborates with Openscapes as an existing mentor, and is an ESIP Committee Member.

Data and Workforce Modernization Is Happening

A common saying when people talk about modernizing is: “Technology is easy; people and social change are difficult.” At NOAA Fisheries, we’ve seen that real change is possible when technical and social infrastructure are prioritized together. Modernization efforts not only include data and technical infrastructure, but also the workforce—providing staff with the necessary skills and collaborative tools to drive these changes. One way we are doing this is the Openscapes Champions program. Openscapes creates a space where members from multiple teams come together to develop joint processes to improve overall project success.

Data modernization at NOAA Fisheries means getting information out faster so people can use this information to meet their needs. It includes data, reports, and analyses and making these more efficient to create, share, and update consistently, relieving staff time. It also makes it more transparent both to the team and people interested in the information.

Over two months in fall 2024, 120 NOAA Fisheries staff tackled projects to improve or restructure data workflows through the Openscapes Program. They made substantial progress on complex workflow transformation goals, such as:

• Transitioning to contemporary data storage solutions;
• Automating large data-heavy reports with programmatic code and version control;
• Developing coordinated science program onboarding and operating procedures that are harmonized across NOAA Fisheries;
• Prototyping cloud computing workflows for new data streams from satellites and ocean modeling.

Program leads organized and taught lessons and skill-building workshops, led small group reflection time, and facilitated coworking sessions. During coworking sessions, participants could brainstorm and make progress on what mattered to them with others working on similar tasks. We can solve problems and complete tasks faster together in a social and collaborative way by reaching outside our individual work circles.

How the Openscapes Champions Program Supports Change

The Openscapes Champions program is a remote-by-design, cohort-based mentorship program. It helps teams reimagine data analysis and stewardship as a cooperative effort, using modern reproducible science skills and tools within collaborative networks. These networks are critical in breaking down silos that lead to inefficiencies and duplication and that prevent peer-learning and problem-solving. The program focuses on open science and data modernization via weekly facilitated coworking sessions. The goal is to develop and strengthen a teaching and learning culture within teams and organizations. This program is not only for scientists; admin, IT, and other management-focused staff have worked on workflow-related projects or to improve management of technical teams.

During the sessions, Champions learned from the successes of past Champions Program participants who transformed the Alaska Marine Mammal Stock Assessment Report process and the California Current Ecosystem Status Report process. Openscapes Mentors from across NOAA Fisheries offices also supported the weekly sessions with their expertise in analyzing big data, software development, and automation.

What NOAA Staff Achieved on Projects Across the Agency

The Champions program is a place for people to solve challenges together. Everyone works on mission-driven tasks for data workflow modernization that can seem (or are) insurmountable alone. Participants presented their work-in-progress efforts from the program in the fifth and final call. The progress these teams made was impressive, representing a deep investment in skills and people.

Onward: Continuing to Modernize Data Workflows

We are working to modernize our data infrastructure and workforce together, and Openscapes is supporting this change. We are building on the momentum and products from previous cohorts since 2020. This momentum has led to collaborations as teams found they had shared tasks and needs, and common skills and workflows to be able to work more easily together. This led to the creation of templates and examples that accelerated the work in the 2024 cohorts. Beyond lines of code, teams and individuals have learned how to run more productive meetings and accomplish work via paired writing and programming. Skills such as shared and co-written agenda notes during meetings and regular coworking sessions, create a dynamic session where peers can ask for and get help. These new collaboration skills are some of the strongest habits that our staff bring to their work.

This work is not happening in isolation. Federal science agencies, including NOAA, have many mandates around open science and open data. The Openscapes Champions program is one of the initiatives helping to enhance transparency, collaboration, and accessibility of federally funded scientific information.

Building the data science workforce at the Fred Hutch - seeing the shift

In Spring 2025 we worked closely with the Data Science Lab (DaSL) staff to lead a second Champions Cohort with the Fred Hutch Cancer Center (and the 25th we’ve led across all groups 🥳). This followed a first cohort 18 months before, in Fall 2023. DaSL staff are part of the Openscapes Mentors community, and collaborated with other Mentors earlier this year in our Mentoring with a Coach Approach series. They have been working on identifying and addressing needs of researchers at the Fred Hutch, one of which is so beautifully shared by Mentor Monica Gerber in her January blog post, Solving the lonely data scientist problem.

The DaSL group has been making progress in supporting data science across the Fred Hutch, helping upskill science teams & making connections across the organization. They have been intentionally growing to meet the community needs, which they identify through working with lab and infrastructure groups across the Hutch. The visible ways we see this are through the increased team size at DaSL, and the amount of support, resources, and software workflows they’ve built in the last eighteen months, including in their regular trainings, wiki, the FH-Data Slack, and the Monday Morning Data Science Newsletter that people look to to learn about trainings and resources (see a great map of these at the bottom of this post). Something especially exciting was seeing Sitapriya Moorthi - a Champion in 2023 and is now a Mentor - really shine in leadership.

During the Spring 2025 Champions Cohort, we saw participants value and grow in their approaches to team culture, sharing work in progress and recognizing that feedback from the community is crucial for making things better. People experienced coworking - many for the first time - in the weeks between Cohort Calls, and saw that coworking with their peers is a tool they can use in different settings (Lowndes et al. 2024). Folks learned they are not alone; DaSL and the Scientific Computing (SciComp) team are there to support staff and are actively welcoming and responding to input. The cohort was able to bridge the gap from knowing that these resources exist, to understanding how the different support systems work at the Fred Hutch and who to go to for what issues. This kind of change in small groups can have outsized impact, particularly when the change is led through example.

The power of “taking stock” to make progress towards better science

In the first Cohort Call, Sitapriya shared her reflections on better science in less time, and how “taking stock” featured prominently (slides | recording). It is about identifying what you have and what you need. It is about being deliberate and writing it down, an often an undervalued exercise. Taking stock is a norm in the “wet-lab” world, in terms of protocols, equipment, reagents, and cells, but not so much in “dry labs” i.e. computational groups. Sita shared how the Openscapes Pathway sheet is one framework for taking stock in a dry lab setting, that helped her shift to better practices digitally, and gave her more control and order in her work. It has enabled continued collaborative upskilling and reproducibility.

From Sitapriya Moorthi’s Better Science presentation; her common misconceptions about Open science that she was able to overcome through working together with Openscapes and DaSL.

Sita has grown as an open data science leader throughout her time at the Hutch; first as a wet-lab scientist, then a dry-lab scientist in the Berger Lab where she was a 2023 Openscapes Champion, and now as a staff scientist with DaSL and 2025 Openscapes Mentor. Sita shared her initial misconceptions about open science; these are the ones that so many other folks have, including Champions participants. Through her own experience in the Openscapes program Sita’s misconceptions were dispelled by having the space to take stock with colleagues across the Hutch, discuss with folks in the Berger Lab and DaSL, and create robust workflows and mindsets that provide resilience and reproducibility to her work.

Cross-learning and seeing shared purpose to support technical infrastructure and people at the Fred Hutch

In this Cohort we taught the core Openscapes Champions lessons, with Sita Moorthi and Ted Laderas (both DaSL Mentors) guest-teaching and sharing examples specific to the needs of biomedical researchers at the Fred Hutch. By design, Openscapes Champions Cohorts bring together people with diverse jobs, experiences, skills, and skill levels. This cohort included people from across the Hutch - from the wet lab, “lonely data scientists”, and infrastructure support teams like DaSL and SciComp. We had 19 people join biweekly Cohort Calls and a small group joining regularly at Coworking in the intervening weeks. Spring 2025 was a hard time to commit for US folks working in the area of health and medicine.

Bringing these people together over ten weeks revealed gaps and opportunities. Participants came looking for advice on how to work better. DaSL Mentors learned that even researchers within the Fred Hutch were not aware of all the resources and technologies available to them. And given pointers to those resources, like the SciWiki or shared high-performance computing infrastructure, how, as a user, do they use them for what they need?

Some explicit user needs examples we heard are that people:

• Look to the Monday newsletter for events, training and news;

• Use the SciWiki as their main reference, but had not realized that it is open even to contributions like an SOP that seems relevant only to a specific to a small group.

• Learned that it’s hard to contribute to open science when you’re brand new to it;

• Want opportunities to cowork to get their own work done, while crossing silos of labs or roles.

Biweekly coworking sessions were really valuable, with regulars attending who had a question each time. People screenshared their work in progress and identified what kinds of “bottleneck tasks” they were seeing in their labs and what kinds of tools or approaches might help avoid the bottleneck strain there. For example, one person committed to working for the first time on a Shiny app. This led to a discussion with a data scientist from another lab about design decisions around the users of the app and the need for a point and click approach. Another person had a metadata decision chart for scientists and wondered how to know if it was working. We talked about how if it was code, you would add tests to it. We can bring that same idea to this metadata decision chart: screenshare to walk someone through it and test it by seeing what questions they have. Many of the lone data scientists who have found each other through this cohort want to continue to use coworking as a tool.

Moving forward: from the lonely data scientist to a community of connected data scientists and superpowered lab members.

Looking ahead, there is continued work to be done around organizational culture and connecting. How do we shift the idea of the lonely data scientist isolated within research labs towards a community of connected data scientists and superpowered lab members?

Over half the participants in this cohort work as individual bioinformaticians embedded in a team or wet lab, saying “I am the only data scientist in my group.” This echoes Monica Gerber’s blog post. These folks are keen to create spaces and a culture not just to get feedback on code, but also learn how the data was generated in the lab. Seeing this as a challenge to be addressed will also help plan forward where to invest energy and attention. If the need is a feeling of belonging in open science and building a sense of community, then making a new tool for analysis or a new channel for asynchronous communication is not the solution. The boundaries of internal lab culture are hard to crack. Researchers in wet labs routinely dedicate time to make and maintain lab notebooks, take inventories, develop protocols, etc. If they do not have the time needed for learning new ways of working or exploring open science resources available to them, then maybe we need to imagine other ways of meeting them where they are. Literally! Maybe it’s time for “Take a DaSL staff member to work day” to meet people in their labs or at their computers, ask what their challenges are, and share the resources that address their specific needs. Researchers can’t bridge this gap alone.

We do have tools we can use and adapt to work on this, we are not alone or starting from scratch. One participant said they “now have a standing ‘database/data management’ meeting; “this is my Seaside Chat to help communicate how I do my job and share my process with my team.” Seaside (or Lakeside) Chats are a place to role model sharing and asking for feedback on imperfect work in progress. People can counter the established norm where folks don’t talk about what didn’t work, and create a place to do that.

The value of cohort coworking in a weekly sense is that people screenshare to get unstuck, say what they’re working on for awareness and accountability, and DaSL sharing Hutch systems and resources that address people’s needs. In a longer term view, Coworking helps change culture by breaking down silos within an organization. DasL had offered coworking before, but the missing piece had been people not knowing what to expect since they had never participated in coworking before. Now people come to weekly DaSL coworking, and can continue to build momentum together!

The new Fred Hutch DaSL resources map below shows the wealth of technical and real live human resources folks can already take advantage of. Going forward, helping people find and use these resources and find each other across the chasms of different research and infrastructure groups is part of the work ahead.

Resources

• openscapes.cloud for NASA and NOAA Fisheries JupyterHub policies and cost reporting.

• 2i2c helps communities build their own interactive computing hub in the cloud with open infrastructure.

• CryoCloud: Accelerating discovery and enhancing collaboration for NASA Cryosphere communities.

• Grafana for monitoring and visualizing usage data.

• AWS Cost Explorer and API docs for managing cloud costs on AWS.

• jupycost: A work-in-progress R package from Openscapes for querying and summarizing JupyterHub cost and usage statistics.

• sixtyfour: An R package for interfacing with AWS APIs, from the Fred Hutch Cancer Center Data Science Lab

• grafana-dashboards: Grafana Dashboards used in our JupyterHubs. Provides Grafana Dashboards as code – very useful for learning how to query Prometheus metrics.
  

Speakers

• Tasha Snow is a co-founder of CryoCloud, a remote sensing glaciologist at ESSIC University of Maryland and NASA GSFC, and was the recipient of the 2023 AGU Open Science Recognition Prize. 

• Andy Teucher is a core Openscapes team member and develops software and cloud infrastructure. He is a biologist turned data scientist, specializing in helping people and organizations build maintainable, reproducible data workflows. Andy is a strong open data and open code advocate, and believes in the value of using and contributing to open-source software.

• Eli Holmes is currently lead of NOAA Fisheries Open Science and in this role, facilitates and runs trainings in computing, data access and statistics for NOAA Fisheries. She is co-lead of the Inter-agency R User Group (federal agencies) and NMFS Openscapes, and other trainings and HackWeeks. 

• Yuvi is a co-founder and tech lead of 2i2c.org, a core member of the JupyterHub team, ex-member of the Wikimedia DevOps team and has been doing open source work for about 17 years in various communities.

• Alex Lewandowski is a research software engineer at the Alaska Satellite Facility. He works on the Science Enabling Services Team. Much of his work focuses on providing support, educational materials, and tools to ASF’s users of SAR data. He has been an Openscapes Mentor since 2022.

We are focused on supporting the missions of the organizations we work with by supporting their staff and scientists 

We’re starting to talk about Openscapes as a mechanism that can rapidly identify problems and bring people together to solve them. The communities below include people within our partner organizations that work together to deliver work aligned with their goals to enable science and society, and are enriched by collaborating with the open science community.

Other good stuff

• Cloud Native Geospatial Conference: on May 1 Julie is keynoting in collaboration with the amazing Eli Holmes (NOAA Fisheries). Eli was invited to speak based on her contributions to the open science geospatial community that are huge and wide-spanning, and include developing novel important infrastructure for science and teaching & building strong communities across NOAA and connected far beyond. Eli is on travel freeze, and it is an honor to support and step in for her, while deeply sad it won’t be her on that stage.

• USGS Community for Data Integration Conference: on April 29 Julie is keynoting to share about open data communities lessons learned across federal and state agencies and academia. We’re excited to learn more about USGS data community progress and needs and meet more folks involved.

• We’ve been experimenting with using our Flywheel each month to reflect together on shared progress and impacts as a team and community and we use that to inform reporting to our partners. 

• jupycost is an R package for JupyterHub cost monitoring and reporting that Andy Teucher is developing as a common cost/usage reporting framework based on needs by NASA and NOAA JupyterHubs, in collaboration with 2i2c. Andy will share more about it at the April 22 Community Call!

• In February we completed the third year of our Pathways to Open Science Program, with 11 speakers and 69 certificates issued to participants! As well as being an important skill- and community-building opportunity for participants, we have been learning through iterating shared leadership roles, designing to include more guest speakers (two in conversation in the same coworking breakout room), and developing a strong sense of community for participants within a drop-in cohort design.

• Julie reflected on the abundance of community at the American Geophysical Union (AGU) Fall Meeting 

• The Openscapes and rOpenSci organizations and communities share values around enabling people doing open and kinder science and many of our activities are complementary. In November, we teamed up for “Getting to know Openscapes” at rOpenSci’s monthly social coworking and the discussion led to a post by Hugo Gruson, Stefanie Butland, and Ruby Krasnow on The Dynamic Relationship of Forks with their Upstream Repository.

• We are making plans for the 2025 ESIP summer conference in Seattle, and looking forward to seeing folks there.

Closing thoughts

Something that has resonated through this work is adrienne marie brown’s idea of fractals – “what we practice at the small scale sets patterns for the whole system.” We see this showing up in small and big ways over short and long time scales as we continue to learn from, with, and for each other. We have been quieter than usual, but we continue to do the work and are here. Thank you for reading, and for all that you do.